Master the theoretical and practical foundation for understanding how large language models process human language through tokenization and word embedding models.
This course provides a technical foundation for understanding how large language models process human language through tokenization and word embedding models. It covers the mechanics of WordPiece tokenization and the architecture of Word2Vec models, specifically the Continuous Bag of Words (CBOW) and Skip-gram approaches. Participants explore the mathematical representation of words as vectors, the role of neural network weights in capturing semantic relationships, and the practical implementation of these models using PyTorch. The course concludes with a demonstration of building a basic mini question-answering system and an introduction to the transition from word embeddings to transformer-based architectures.
Topics Covered:
5 activities • 31 videos • duration: 6h 7m

Robotics and AI Engineer
Nawas Naziru Adam currently works as a Principal Robotics and AI Engineer. He obtained his Bachelors Degree in Computer Engineering and a Master's Degree in Information Technology. He had the opportunity to work as a Lead Data Scientist. He has executed a number of projects involving AI, Robotics and Embedded Systems. I love breaking down complex concepts and thus, on a mission to find and distribute the keys to knowledge, learning and mastery.
Professional development requirements vary by profession, country, and regulatory body. Before enrolling, review the course details and confirm that the content, learning format, and any listed credits meet your specific requirements. You are responsible for confirming eligibility with your regulator or professional association and keeping the records required for reporting.